In the past few years, ambiguous results have been published in the literature concerning the influence of monetary policies on real variables. The objective of this paper is to test if the effects of monetary variables (here several interest rate spreads) on real variables (growth in real GDP, unemployment, changes in the industrial production index) can be modelled by a nonlinear function of the "smooth transition auto-regressive" (STAR) form. Our results, generated with United States data, confirm the existence of such functions which can explain why results obtained with linear models lack robustness.
AFTALION, F. (1995). Asymmetries in Monetary Policies : The Star Model. In: Forecasting Financial Markets. Imperial College, pp. 1-21.